Lower limb bone model construction method, device, electronic equipment and storage medium

Abstract

The embodiment of the present invention discloses a lower limb bone model construction method, device, electronic equipment and storage medium. The method includes: obtaining a sequence of original scanned image frames obtained by continuous scanning of human lower limbs; For human body parts, segment each of the original scanned image frames to form at least two segmented regions corresponding to different human body parts; determine a boundary segmentation strategy according to the human body parts corresponding to each segmented region; according to the boundary segmentation strategy, Skeletal boundary segmentation is performed on the image frames in the segmented region. Through the technical solutions of the embodiments of the present invention, the technical effect of accurately and rapidly constructing lower limb bone models is achieved.

Description

technical field [0001] The embodiments of the present invention relate to the field of medical technology, and in particular to a method, device, electronic equipment and storage medium for constructing a skeleton model of a lower limb. Background technique [0002] Artificial knee replacement is one of the important means of treating knee diseases. Before performing artificial knee replacement, it is necessary to fully grasp the morphological and structural characteristics of the patient's lower limb bones. Therefore, in order to accurately extract the features of the human lower limb from the femoral head to the ankle joint area, it is necessary to accurately segment and reconstruct the above bony structures. [0003] Artificial knee arthroplasty usually acquires bone data of the lower limbs through sequential medical images collected by computerized tomography (CT). However, due to the inclusion of thousands of CT images, manual segmentation of bone structures is diffic...

AI Technical Summary

Problems solved by technology

However, traditional segmentation methods include filtering, automatic thresholding, morphological processing, etc. It is difficult to accurately distinguish the segmentation boundaries of bones and other tissues through traditional segmentation methods, and there is a problem of low segmentation accuracy; deep learning methods require a large amount of labeled data. To train and test the deep learning model, however, the labeled data of human lower limb bones is difficult to obtain, and there are individual differences in human lower limb bones, and there are still problems of low segmentation accuracy and speed

Images